Tree transpiration is a critical process of the water cycle and its observation and quantification are essential to better understand terrestrial ecosystem dynamics. While sap flow measurements provide direct estimates of individual tree transpiration, their reliance on empirical calibration and a rather high energy consumption (due to heat generation) can be strong limiting factors. The self-potential (SP) method, a passive geophysical approach, presents a promising alternative for assessing transpiration rates, although the electrophysiological processes driving SP signals in trees remain underexplored (Fig. 1). This study presents a year-long monitoring of SP and sap velocity in three tree species at three Critical Zone Observatories different in a Mediterranean climate (from OZCAR and ICOS research infrastructures). Using wavelet coherence analysis and variational mode decomposition, our findings reveal strong coherence between SP and sap velocity at diurnal time scales, with coherence diminishing and phase shifts increasing under higher water supply conditions. Electrokinetic coupling coefficients, derived from linear regression between SP and sap velocity variations, align with values typical of porous geological media. During dry seasons, the electrokinetic effect dominates SP signals, suggesting its potential as a tool for evaluating transpiration rates. This research demonstrates the feasibility of integrating geophysical techniques into long-term observations of ecohydrological systems to better understand plant-water interactions in the Critical Zone.

Introduction Modelling the hydrological processes in mountainous areas is particularly challenging due to the strong heterogeneity of the underground medium in terms of hydrological properties, and the lack of groundwater observations. Here, we show how geophysical observations provide key information on the geometry of hydrofacies, the estimate of hydrological properties and the monitoring of the groundwater to study the critical zone in the Strengbach mountainous headwater catchment. Studied Site The OHGE (Observatoire Hydrogéochimique de l’Environnement) is a headwater catchment of 0.8 km² that lies on a granitic bedrock (Pierret et al. 2018). This observatory corresponds to the Strengbach catchment and is part of OZCAR, the French network of critical zone observatories. The OHGE is located in the Vosges mountains (northeastern France) with altitudes varying between 880 m and 1150 m (Fig. 1). The catchment topography shows steep slopes of 15° in average that reach up to 30° locally. Learning from scattered geophysical data Meteorological and hydrological data are monitored since 1986 and six boreholes provide the distribution of geological facies at depths of 50 to 120 m (Chabaux et al. 2023). In addition, electrical resistivity and seismic refraction tomographies were acquired to estimate variations in soil and saprolite thickness. These data show soil thickness varying from 50 cm to 5 m, and saprolite thickness ranging from 1 to 16 m (Lesparre et al. 2024). The electrical resistivity tomographies also underline the spatial distribution of the geological facies, as one slope of the catchment shows significantly higher resistivity values than the other (Lajaunie et al. 2024). Despite the relatively thin saprolite, magnetic resonance soundings detected groundwater above the noise level, and revealed a region with higher water content (Lesparre et al. 2020). Gravity data acquired across the whole catchment shows that the method has the sensitivity to distinguish areas with distinct water storage dynamics (Chaffaut et al. 2022). In particular, a region upstream the Strengbach stream exhibits the highest values of water content and the largest variations in water storage. We are using the geophysical data to develop a catchment-scale hydrogeophysical inversion. Hydrofacies geometries will be derived from the tomographies, and local hydrogeophysical experiments will be interpreted together with direct observations to estimate the range of hydrological properties. That information will serve as prior information of the inverse problem that will assimilate magnetic resonance and gravity data to complete piezometer and flow rate data.

Due to their high resolution and near real-time capabilities, borehole strainmeters are a critical complement of satellite-based geodetic systems [GPS/global navigation satellite system (GNSS)] and satellite interferometry. However, commercial strainmeters remain expensive and only provide the three horizontal components of the strain tensor. We propose a novel design able to detect the six components of the 3-D strain tensor at relatively low cost. Here, we embed six compliant elastic gauges in a sphere in order to evenly sample space directions. Each gauge exhibits an amplification ratio of similar to 30. Interrogated by Fabry-Perot interferometers illuminated by a single laser diode (LD) through a multichannel fiber cable, these opto-mechanical systems exhibit a resolution of similar to 89 pm/root Hz over a dc - 500-Hz bandwidth, ultimately providing a strainmeter resolution <1 nanostrain. A supplementary interferometric device is also implemented to detect and correct nongeometrical optical phase changes due to pressure and temperature variations. An original building method allowed us to mold fibered cement around a thin sphere equipped by the optical strain gauges. Moreover, an adjustable pressure device is integrated to allow in situ calibration in order to correct the 3-D strain tensor from borehole and cement heterogeneities. This cost-effective strainmeter was successfully installed in November 2023 in a 30-m deep borehole at the Larzac Observatory in the French Massif Central.

Within the critical zone, rock weathering plays a fundamental role in controlling subsurface water storage and transfer. In karstic carbonate settings -particularly sensitive to dissolution and erosion -characterizing the geometry and properties of weathered rocks is essential. This study proposes a conceptual model of weathered zones in the subsurface of the critical zone, based on an integrated imaging framework. The study site is in a dolomitic karst area (Larzac observatory, Southern France). First the properties of weathered and unweathered rocks are analyzed on rock samples using petrological observations and petrophysical measurements. Weathered rocks appear as unconsolidated, highly porous material, resulting from dissolution at the grain boundaries of the dolomite without any significant mineralogical change from the unweathered rock. Various geophysical methods (GPR, Gravimetry, ERT, SRT) are used to identify and image the weathered areas in subsurface. The weathered rocks form numerous corridors (several meters long) both at the surface and at depth, aligned with the fractures and known caves. By integrating multidisciplinary data from samples and field, we infer the potential presence of water in these weathered zones. This leads to a refined conceptual model of karst subsurface architecture, where weathered rocks may serve as preferential reservoirs for groundwater within the critical zone.

The objective of this work is to investigate infiltration processes, flows on both saturated and unsaturated zones, and quantify their contributions to spring discharge, in a binary karst aquifer (recharged by the karst massif, as well as sinking streams). The study site is the Verneau karst aquifer located in the Jura Mountains (France). The recharge area covers 15 km², with half consisting of marl outcrops, where water enters through five losses (localized infiltration), and the other half consisting of limestone massif covered by a soil layer (diffuse infiltration).First, to analyze the spatial variability of flows generated by diffuse and localized infiltration at various depth, we performed hydrochemical analyses in soil lysimeters, caves and at the spring, conducting to characterize the physico-chemical end-members of the various compartments: soil layer, unsaturated zone in the karst massif and in the conduit network, saturated zone. Second, we used continuous high-frequency (1hour) time series (3 years) of semi-conservative tracers (electrical conductivity and nitrate concentrations) to characterize discharge response throughout the seasonal cycle. A End-Member Mixing Analysis (EMMA) was conducted on 40 flood events to determine the contribution of the infiltrations types to spring discharge. Our results show that the localized infiltration shows a relatively homogeneous spatial signal, characterized by low values of electrical conductivity and nitrate content. Diffuse infiltration is spatially variable due to anthropogenic activities and contrasted residence time within the massif. Results of the EMMA method reveal that during flood events, approximately 1/3 of spring discharge comes from localized infiltration, while the majority comes from diffuse infiltration and pre-event water stored in the massif. A seasonal variability is evidenced in link with lower stream losses and storage in the unsaturated zone. A hydrogeological conceptual model is finally proposed, allowing us to discuss the origin of spring waters, given new elements on drivers controlling infiltration modalities, and the role of transfer and storage in the unsaturated zone.

Forests cover almost one third of the Earth's land area and are central in the carbon and water cycles. Soil water availability is one of the most important factors regulating transpiration, biomass production and plant species distribution in ecosystems. The carbon and water cycles are closely linked and so understanding the functioning and evolution of forest environments and their relation to subsurface structure and water availability is essential to improve understanding of the water cycle under a changing climate. Studying the forest subsurface is a challenge because of its heterogeneous nature and difficult accessibility. Traditional approaches used by ecologists are also often point measurements that have a low spatial representativity. Near-surface geophysics offers a wide range of methods to characterize the spatial and temporal variability of subsurface properties and associated processes in a non-destructive and integrative way. Geophysical methods allow us to obtain new information that complements ecophysiological methods to better understand ecosystem functioning, and in particular processes linked to ecohydrology. The use of geophysical methods in forests is growing, both by geophysicists seeking to apply their tools to more complex environments, and by ecologists seeking to better characterize their experimental sites. One of the major applications and assets of geophysics in forests is to quantify and monitor water stocks and dynamics. For example, geoelectrical monitoring can be used to assess the distribution and spatial variations of water content in the subsoil. In this work, we show the example of a recently developed ensemble approach to quantitatively relate electrical conductivity monitoring and the distribution and dynamic of water in forest soils. We believe that such interdisciplinary advances can help us improving the quantitative assessment of forest responses to the environment and their adaptation to climate change.

Plant transpiration is a crucial process in the water cycle, and its quantification is essential for understanding terrestrial ecosystem dynamics. While sap flow measurements offer a direct method for estimating individual tree transpiration, their effectiveness may be limited by the use of point sensors; species-specific calibration requirements; and baseline uncertainties, particularly the assumption of negligible nighttime flow, which may not always hold. Self-potential (SP), a passive geophysical method, holds potential for constraining transpiration rates, though many questions remain regarding the electrophysiological processes occurring within trees. In this study, we continuously measured tree SP and sap velocity on three tree species for 1 year in a Mediterranean climate. Using wavelet coherence analysis and variational mode decomposition, we explored the empirical relationship between tree SP and transpiration. Our analysis revealed strong coherence between SP and sap velocity at diurnal timescales, with coherence weakening and phase shifts increasing on days with higher water supply. We estimated electrokinetic coupling coefficients using a linear regression model between SP and sap velocity variations at the diurnal scale, resulting in values typically found in porous geological media. During dry seasons, the electrokinetic effect emerges as the primary contribution to tree SP, indicating its potential utility in assessing transpiration rates. Our results emphasize the need for improved electrode configurations and physiochemical modelling to elucidate tree SP in relation to transpiration.

On Earth, impact structures are rare in intertropical zones. Here we evaluate the 35-40 km diameter Velingara depression in Senegal as a candidate impact structure. The depression has a topographic relief of only 50 m and is essentially buried under modern sediments. For the first time, potential-field ground observations and a microstructural survey of detrital zircons for shock features were conducted at Velingara. A 15 mGal gravity low in a 10 km diameter central anomaly was found, while significant (100 nT) amplitude magnetic field anomalies were observed in the depression. Forward modeling was used to explain the crustal sources of the anomalies, arguing in favor of the presence of a buried impact structure. An electron backscatter diffraction survey of circa 5,000 detrital zircons from a sample of modern alluvium identified a single granular zircon; orientation data for the grain does not record the former presence of reidite, but does closely resemble grains with highly dispersed neoblast orientations that have only been reported from impact settings. Age determination by LA-ICP-MS indicates the unusual granular zircon recorded an age of ca. 550 Ma. Results of the geophysical data and zircon microstructural survey reported here are consistent with, but do not yet prove, an impact origin hypothesis for the Velingara structure. The difference in amplitude between the diameter of the depression, and that of the gravity anomaly remains puzzling. Several impact scenarios are considered, which include the subsequent evolution via surface processes of a 10-km crater into a larger depression, or a large-scale (d = 40 km) complex impact crater with a peculiar geophysical anomaly limited to its central region.

AbstractIdentification of preferential flow‐paths, such as fractures, is required for various issues in geosciences. When chemicals are injected into the subsurface, monitoring the resulting structural and chemical changes remains a challenge. The ability of geophysical tomography to tackle this problem is not fully explored due to the lack of numerical methods suitable for modeling narrow structures. We explore how discrete representation of preferential flow‐paths provides innovative ways to invert electrical resistivity data collected during reagent injection at a contaminated site. The data set is inverted with a scheme where a new fracture is added at every iteration. This allows identifying newly created narrow conductive structures from the field data collected before and after injection. Fracture location remains overall consistent despite using different starting points for the fracture search. A prior constraint on fracture length improves convergence. These results show the potential of discrete inversion for identifying narrow structures from electrical resistivity monitoring.

The Gravimetry Specific Action (ASG) of the French Research Infrastructure (RI) Epos-France is based on a mutualized national instrumental park offering scientific teams a range of measurement resources and services for the gravity field studies. It is driven by an office which aims to manage all activities in the field and runs the park. The range of instruments includes absolute and relative gravimeters of different classes and technologies. It allows covering a vast array of applications with which, among other things, it maintains the French observation networks, the “measurements” component of the National Gravimetry Observation Service (SNO) in France.

Karst systems represent an important carbon and freshwater reservoirs. Although karst systems have been studied for many years, a new paradigm has emerged that suggests some of them could be formed by ghost-rock processes (Dubois et al. 2014). Contrarily to the classical total karstification, ghost-rock karstification leaves in place a weathered rock, called the ghost-rock, that can constitute a microbial habitat (Spilde et al. 2005). The first results of a geomicrobiological study of the Sterkfontein’s cave system in South Africa show that these ghost-rocks are mainly composed of iron and manganese oxides mixed with organic matter of putative microbial origin (Pisapia et al. in prep). To further understand the microbial community inhabiting these ghost-rocks, its specificity compared to groundwater, and its functional impact on the karst system of Sterkfontein, a metagenomic analysis from both ghost-rocks and groundwater samples was performed. It was completed by laser microdissection of the microorganisms attached to the mineral particles, followed by whole-genome amplification and transmission electron microscopy to analyze both the nature of the mineral particles and the microorganisms associated with them. The results highlight the differences in community between these two environments (with higher abundance of Actinobacteriota and Acidobacteriota in ghost-rock samples compared to ground water in particular), and suggest a high importance of microbe-minerals interactions in the ghost rocks, through metallophores production and extracellular electron transfer processes between bacteria and metallic ions.

The objective of this study was to evaluate the relevance of using excess air (EA) for the characterization of drain/matrix exchange in karst systems using a rainfall discharge model coupled with the simulation of EA measured at the outlet of the studied system. The conceptual model assumes a linear relationship between the formation of EA and the increase of hydrostatic pressure in the capacitive part of the aquifer. The simulated EA at the spring consists of the mixing of water circulating in the different compartments of the aquifer, with their own EA signature. The analysis is performed taking as an example the Durzon karst system (Larzac, France). The modeling is applied using daily rainfall discharge time series and 18 EA measurements at the main outlet of the karst system within 3 hydrological cycles. The main modeling results show that EA variations measured at the karst spring can be explained by recharge processes and exchange between conduit and matrix. EA measurements at the spring thus contain valuable information about the flow dynamics within the aquifer. Furthermore, results show that the use of EA measurements, despite their sparse temporal resolution, allows for reducing uncertainties in the estimation of some parameters of the reservoir model used for the simulation of karst spring discharge.

The eLTER H2020 process aims at developing an European ecosystem research infrastructure and at creating long term observations platforms. In this context, we proposed a platform called P3M (Mediterranean Plain, Piedmont and Plateau) located in the Mediterranean basin that is considered as a hot-spot of both biodiversity and climate change. P3M provides an observation system of several contrasted socio-ecosystems representative of the Mediterranean diversity from local to territorial scales. P3M will study the connection between strong hydro-pedo-climatic gradients and biodiversity that are subjected to an increasing variability of hydroclimatic processes, while being strongly connected to anthropic drivers. P3M is structured around a toposequence of four different agro-ecosystems where the natural and cultivated sphere are highly connected : (i) Cevenol medium-size mountains characterized with hillside forest ecosystem and extensive agriculture on the valley bottom, (ii) High-altitude plateau with an agropastoralism-dryland system and a karstic hydrogeology reagent to precipitations that highly influences the quality and quantity of the water ressource on the entire platform, (iii) Cevenol piedmont natural forest and garrigue ecosystems, (iv) Herault watershed hilly plains organized around cultivated plains and mainly viticultural activities. This scientific collaboration will focus on i) analyzing the hydro-eco-sedimentary dynamics in between each ecosystems but also throughout the entire toposequence, ii) identifying the interactions between those dynamics and human activities within the different ecosystems iii) predicting the evolution of mediterranean landscape facing climate change and providing adapting solutions. To face these challenges, a transdisciplinary approach is necessary to better understand the connections between the increasing pressure of human activity, climatic phenomena, biophysic processus and biodiversity. P3M will answer European environmental scientific issues by ensuring the stability, the availability and the European standardisation of abiotic, biotic and social datas on a long term scale but also the reception of foreigners scientists. This platform gathers four OZCAR labelised observation networks (OMERE, Larzac Observatory, OHM-CV, MEDYCCYS) but also includes ANAEE and ICOS sites. Non-academics partners will take part in this project, for example the national and regional park or the local watershed facilities, to integrate the existing data but also to facilitate the transfer of research to local policies.

With no significant topographic expression and limited bedrock exposure, the similar to 10 km diameter Karla impact structure (Tatarstan, Russia) is poorly known. The age of the impact is also poorly constrained stratigraphically to between 4 and 60 Ma, even if an upper Miocene age is more likely. Targeted gravity and magnetic field surveys were conducted over Karla to explore its size and structure in 2019. Bouguer gravity anomaly data reveal a central positive (+2 mGal) peak similar to 2 km in diameter surrounded by a concentric negative (-1 mGal) anomaly extending to similar to 3 km radius; a more irregular, outward-decreasing (+1 to -1 mGal) positive anomaly extends to 6-8 km radius. A complex impact structure with diameter of 8-10 km is consistent with the Bouguer anomalies. Magnetic field data show 1 to several km-wavelength anomalies with amplitude variation from +150 to -150 nT and little concentric structure, although the impact feature broadly corresponds to a magnetic low with a weak central high. A 2-D numerical model of the structure was built using these potential-field data and petrophysical properties measured on collected samples. It confirms a central uplift composed of Paleozoic sediments and Archean crystalline basement up to 1 km of depth. A 500 m deep collapsed disruption cavity filled by breccia and lacustrine deposits accounts for the Bouguer negative ring. The reversely polarized and weak central magnetic anomalies are controlled by the geometry of the crystalline basement associated with the deformation during the central uplift.

Evapotranspiration (ET) is a major component of both the hydrological cycle and the surface energy balance. Furthermore, ET is strongly linked to the primary production of natural and cultivated vegetation covers. Therefore, obtaining spatialized estimates of ET is of paramount importance in Mediterranean areas, submitted to hot and dry summers, all the more so as climate change is expected to worsen the water deficit in this region. In the framework of the preparation of the TRISHNA satellite mission, the objective of this study was (1) to produce maps of ET over a small Mediterranean region from high resolution satellites, and (2) compare these satellite estimates with local measurements of ET from flux towers. The studied area was located in the Hérault river area, south of France. During the period 2013 - 2019, 63 clear sky Landsat 7 and 8 images were collected and processed, having spatial resolutions of 60 and 100 meters, respectively. Maps of ET were generated using the EVASPA processing tool (Gallego-Elvira, 2013), with various methods for estimating the soil heat flux and the evaporative fraction. These satellite estimates were compared to those measured by long term flux towers installed on three biomes representative of Mediterranean landscapes: Puechabon (Quercus ilex forest on a rocky soil), Larzac (grassland on a limestone plateau) and Roujan (vineyards in the plain). In the estimation of ET from satellite images, intermediate variables (albedo, net radiation, soil heat flux, surface temperature) were first compared to those measured locally, when available. Finally, instantaneous and daily satellite estimates of ET were compared to the local measurements. Depending on sites and EVAPSA methods, the RMSE of instantaneous estimates of ET ranged between 39 and 202 W.m-2. The RMSE of daily estimates of ET ranged between 0.96 and 1.82 mm.day-1. Future work will be conducted to analyze the effect of air temperature variations, induced by altitude variations, on ET satellite estimates. This will allow to extend this study to larger regions of southern France.

&lt;p&gt;Characterizing and monitoring water flow in the critical zone is of uttermost importance to understand the water cycle. Water link several process within critical zone from aquifer recharge and solute transfer to eco-hydrology, many eco-systemic services and biogeochemical reactions. However, the &lt;em&gt;in situ&lt;/em&gt; quantification of water flow is technically challenging using traditional hydrological methods and numerous gaps of knowledge remain. The self-potential (SP) method is a passive geophysical method that relies on the measurement of naturally occurring electrical field. One of the contributions to the SP signal is the streaming potential, which is of particular interest in hydrogeophysics as it is directly related to both the water flow and porous medium properties. Unlike tensiometers and other point sensors, which use the measurement of state (e.g., matric pressure) at different locations to infer the intervening processes, the SP method measures signals generated by dynamic processes (e.g. water movement). However, the amplitude of the SP signal depends on multiple soil properties which are dependent to soil type, moisture content, and water chemistry (composition and pH). During the last decades, many models have been proposed to relate the SP signal to the water flow. In this contribution, we will present a soil-specific petrophysical model to describe the electrokinetic coupling generated from different water fluxes in the critical zone: rain water infiltration and water uptake from tree-roots. We tested a fully coupled hydrogeophysical approach on a large SP dataset collected in a two-dimensional array at the base of a Douglas-fir tree (&lt;em&gt;Psuedotsuga menziesii&lt;/em&gt;) in the H.J. Andrews Experimental Forest in central Oregon, USA. We collected SP measurements over five months to provide insight on the propagation of transpiration signals into the subsurface with depth and under variable soil moisture. The coupled model, which included a root-water uptake term linked to measured sap flux, reproduced both the long-term and diel variations in SP measurements, thus confirming that SP has potential to provide spatially and temporally dense measurements of transpiration-induced changes in water flow. Similar set-ups are being installed on several test-sites of the French Critical Zone observatory network, OZCAR: Larzac, LSBB, Strengbach. This will allow us to test the approach under different climatic conditions, different soil types and in different ecohydrological systems.&lt;/p&gt;

Fractures are fundamental discontinuities to understand the exchange and the dynamic of water, matter and energy between the different parts of the critical zone. But except from boreholes and direct observation at rocks outcrops, imaging fracture network from the surface remain a challenging task in geosciences. Classical geophysical methods and their associated imaging methods are in a large extend adapted to retrieve the properties and (part of) the heterogeneities of the rock matrix and not of the discontinuities. The aim of the study is to present a stochastic imaging method based on the direct simulation of Electrical Resistivity Tomography (ERT) dataset from Discrete Fracture Network (DFN). ERT is one of most used geophysical method in critical zone studies. And one can find numerous examples in literature of interpretation of ERT images as fracture zone, even if the inversion do not have the resolution to retrieve the fracture geometry or properties. First the direct approach is described (from Roubinet et al., 2014) to simulate ERT dataset. Then from simple cases such one vertical or horizontal fracture, the inversion is detailed with emphasis on the convergence rate, resolution and depth of investigation. Further step towards the inversion of realistic DFN are discussed: how prior information can be used to enhance the convergence rate and how the non-unicity of the solutions can be retrieved from the stochastic inversion scheme. The inversion of DFN from ERT dataset is a first step towards coupled inversion such as pumping tests or tracers tests and ERT.

Estimating evapotranspiration (ET) is a primary challenge in modern hydrology. Hydrogravimetry is an integrative approach that provides highly precise continuous measurement of gravity acceleration. However, large-scale effects (e.g. tides, polar motion, atmospheric loading) limit the fine time-scale interpretation of this data and processing leads to residual signal noise. To circumvent this limitation, we exploited the difference between two superconducting gravimeters located 512 m apart on the same vertical. The difference calculation makes it possible to remove shared large-scale effects. Daily variation of this gravity difference is significantly correlated with daily evapotranspiration as estimated using the water balance model SimpKcET (p-value = 4.10-10). However, this approach is effective only during rain-free periods. In the future, comparison with direct ET measurements (e.g. eddy-covariance, scintillometer) may confirm and strengthen our interpretation. Improved hydrogravimetric data processing will allow to extend this approach to other experimental sites equipped with a single superconducting gravimeter.

More than 80 million m3 per year are pumped into the Roussillon plain coastal aquifer, covering 850 km² and located between the Pyrenean massif to the west and the Mediterranean Sea to the east, south of France. This is a multilayer aquifer of more than 350 m thick, made up of sandy layers embedded in low-permeability clayey material from the Pliocene and topped by alluvial formations from the Quaternary. Its groundwater resource is primarily used for the supply of drinking water, but also contributes to the irrigation of some 13,000 hectares. For more than 40 years, this aquifer has been undergoing a general decline in its piezometric level due to pumping and water demand is expected to increase (growing irrigation areas and climatic demand). Moreover, given its flat topography, the Roussillon plain is likely to suffer sea water intrusions and marine submersion, due to the sea level rise, which could reach 1 m by 2100. This context shaped the Dem'Eaux Roussillon project, which brought together nearly ten partners from the Occitanie region (research units, consultancies and local authorities). Its objective was to characterise the behaviour of the groundwater resource in this aquifer, in order to be able to project its future situation, in the context of climate change, rising sea levels (risk of saline intrusion) and changes in water use. A detailed characterisation of the geological reservoir highlighted the need to consider the offshore extension of this coastal aquifer. The analysis of the piezometric evolution at the scale of the Roussillon plain over the last 50 years allowed the spatialized characterization of the hydrodynamic parameters and the understanding of the vertical drainage processes that control the hydraulic equilibrium between the Quaternary and the Pliocene water tables. Two high-resolution hydro-geophysical observatories have been set up to quantify these processes and improve understanding of saline intrusions processes. Finally, a conceptual model presenting the main features of the main processes controlling the groundwater evolution and the sea water intrusion risk was obtained ready to launch a numerical modelling work.